# US 11,432,761 B1 · Method and server for dementia test based on voice question and answer using artificial intelligence call

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,432,761 B1
- Document ID: US11432761B1
- Status: Granted
- Kind: B1
- Application number: 17/749,867
- Filed: 2022-05-20
- Granted: 2022-09-06
- Earliest claimed: 2021-12-08
- Assignees: Sevenpointone Inc.
- Applicants: Sevenpointone Inc.
- Inventors: Hyeonjun Lee; Juyeong Yoo
- Web page: https://patentgrep.com/patent/US11432761B1

## CPC

- A61B 5/4088: {Diagnosing of monitoring cognitive diseases, e.g. Alzheimer, prion diseases or dementia}
- A61B 5/0022: {Monitoring a patient using a global network, e.g. telephone networks, internet}
- A61B 5/4803: {Speech analysis specially adapted for diagnostic purposes}
- A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems}
- G06F 40/279: Recognition of textual entities
- G10L 15/187: Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L 15/30: Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
- G10L 25/66: for extracting parameters related to health condition (detecting or measuring for diagnostic purposes A61B5/00)
- H04M 3/42221: {Conversation recording systems (at the subscriber's set H04M1/656)}
- A61B 5/6898: {Portable consumer electronic devices, e.g. music players, telephones, tablet computers}
- G10L 15/1822: {Parsing for meaning understanding}
- G10L 2015/223: {Execution procedure of a spoken command}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)

## Abstract

The present disclosure relates to a method and a server capable of performing a dementia test without the need for a dementia test subject to personally visit a test center and capable of automatically performing a dementia test by an artificial intelligence, not an individual. According to an embodiment of the present disclosure, a dementia test may be performed without the need for a dementia test subject to personally visit a test center, thereby improving the effectiveness of a dementia test by filtering patients who need to visit the test center and proceed with a follow-up test, and a dementia test may be automatically performed by an artificial intelligence, not an individual, thereby saving the time required for the dementia test and costs such as labor cost for the test.

## Claims

### Claim 1

A dementia test method based on an artificial intelligence telephone call performing a dementia test based on question and answer using the artificial intelligence telephone call performed on a server, the method including:
providing guidance to a user terminal to which an artificial intelligence telephone call is connected;
conducting a first test, which is an initial familiarization test, in order to increase a test subject's understanding of a test progress method;
conducting a second test, a main test, used to determine a presence of dementia or a dementia symptom level of the test subject;
analyzing content of the first test;
analyzing content of the second test; and
transmitting a test result to the user terminal,
wherein the conducting the first test includes:
providing a first question for requesting an answer to a first topic to the user terminal for a first time period; and
acquiring a first answer from the user terminal;
wherein the conducting the second test includes:
providing a second question for requesting an answer to a second topic, different from the first topic, to the user terminal for a second time period, longer than the first time period; and
acquiring a second answer from the user terminal;
wherein the analyzing the content of the first test includes:
calculating an understanding value, which is a value numerically quantifying a degree of the test subject's understanding of the test progress method based on the first answer; and
when the understanding value is lower than a preset value, determining that the test subject does not understand the test progress method, and guiding the test progress method without conducting the second test or scheduling a test thereafter; and
wherein the analyzing the content of the second test includes:
converting the second answer into text data;
extracting at least one word corresponding to the second topic from the text data;
calculating a language fluency value based on at least one word corresponding to the extracted second topic;
comparing and analyzing the language fluency value with a preset reference value; and
when the language fluency value is lower than the preset reference value, determining that the test subject is a subject for a follow-up test.

### Claim 2 (depends on 1)

The dementia test method of claim 1, wherein the first time period is 10 seconds or more and 30 seconds or less, and the second time period is 30 seconds or more and 60 seconds or less.

### Claim 3 (depends on 1)

The dementia test method of claim 1, wherein the calculating of the language fluency value includes an operation of applying and then scoring an adjusted score criterion to at least one of a total number of words, a number of words in a first half, a number of words in a second half, a number of letters per word, a number of category changes, a number of words by category, and a number of repeated words.

### Claim 4 (depends on 3)

The dementia test method of claim 3, wherein the preset reference value is set differentially for each of test groups classified based on auxiliary information including at least one of gender, age, education level, and a number of people living together.

### Claim 5 (depends on 1)

The dementia test method of claim 1, wherein the second test content analysis operation further includes an operation of analyzing a change in a speech style based on the second answer and existing test data of the user terminal, in the case in which the user terminal has a test history, wherein the speech style includes a speech speed and pronunciation accuracy.

### Claim 6 (depends on 1)

The dementia test method of claim 1, further including:
an operation to confirm an intention to conduct a test;
an auxiliary information collection operation; and
an operation of classifying a test group based on the auxiliary information.

### Claim 7 (depends on 6)

The dementia test method of claim 6,
wherein the auxiliary information collection operation includes:
an operation of providing an auxiliary information question asking any one of gender, age, education level, and a number of people living together to the user terminal; and
acquiring an auxiliary information answer from the user terminal, and
wherein the operation of classifying the test group includes:
an operation of converting the auxiliary information answer into text data; and
an operation of classifying the test group based on the text data.

### Claim 8 (depends on 1)

A computer-readable recording medium in which a computer program for executing the dementia test method based on the artificial intelligence telephone call of claim 1 is stored in combination with a computer, which is hardware.

### Claim 9

A dementia test server based on an artificial intelligence telephone call performing a dementia test based on question and answer using the artificial intelligence telephone call, the server including:
a communication unit communicating with a user terminal to provide an artificial intelligence telephone call, transmitting a question and test result to the user terminal, and receiving an answer from the user terminal;
a dementia test progress unit providing guidance to the user terminal, conducting a first test, which is an initial familiarization test, in order to increase a test subject's understanding of a test progress method, conducting a second test, a main test, used to determine a presence of dementia or a dementia symptom level of the test subject, wherein the first test is conducted by providing a first question for requesting an answer to a first topic to the user terminal for a first time period and acquiring a first answer from the user terminal, and the second test is conducted by providing a second question for requesting an answer to a second topic, different from the first topic, to the user terminal for a second time period, longer than the first time period, and acquiring a second answer from the user terminal; and
an analysis unit analyzing content of the first test by calculating an understanding value, which is a value numerically quantifying a degree of the test subject's understanding of the test progress method based on the first answer, and when the understanding value is lower than a preset value, determining that the test subject does not understand the test progress method, and guiding the test progress method without conducting the second test or scheduling a test thereafter, and analyzing content of the second test by converting the second answer into text data, extracting at least one word corresponding to the second topic from the text data, calculating a language fluency value based on at least one word corresponding to the extracted second topic, comparing and analyzing the language fluency value with a preset reference value, and when the language fluency value is lower than the preset reference value, determining that the test subject is a subject for a follow-up test.

### Claim 10 (depends on 9)

The dementia test server of claim 9, wherein the first time period is 10 seconds or more and 30 seconds or less, and the second time period is 30 seconds or more and 60 seconds or less.

### Claim 11 (depends on 9)

The dementia test server of claim 9, wherein when calculating the language fluency value, the analysis unit is configured to apply and then score an adjusted score criterion to at least one of a total number of words, a number of words in a first half, a number of words in a second half, a number of letters per word, a number of category changes, a number of words by category, and a number of repeated words.

### Claim 12 (depends on 11)

The dementia test server of claim 11, wherein the preset reference value is set differentially for each of test groups classified based on auxiliary information including at least one of gender, age, education level, and a number of people living together.

### Claim 13 (depends on 9)

The dementia test server of claim 9, wherein when analyzing the content of the second test, the analysis unit is configured to analyze a change in a speech style based on the second answer and existing test data of the user terminal, in the case in which the user terminal has a test history, and
wherein the speech style includes a speech speed and pronunciation accuracy.

### Claim 14 (depends on 9)

The dementia test server of claim 9, wherein the analysis unit is configured to:
confirm an intention to conduct a test;
collect auxiliary information; and
classify a test group based on the auxiliary information.

### Claim 15 (depends on 14)

The dementia test server of claim 14, wherein when collecting the auxiliary information, the analysis unit is configured to:
provide an auxiliary information question asking any one of gender, age, education level, and a number of people living together to the user terminal; and
acquire an auxiliary information answer from the user terminal, and
wherein when classifying the test group, the analysis unit is configured to:
convert the auxiliary information answer into text data; and
classify the test group based on the text data.